Free AWS Certified AI Business Strategist (AIB-C01) Practice Questions

Sample AIB-C01 Practice Questions

Browse all 20 free AWS Certified AI Business Strategist practice questions below.

1. A retail chain is reviewing three initiatives at a leadership meeting. The first is a demand forecasting system trained on five years of store sales history. The second is an assistant that drafts product descriptions from a short brief. The third is a rules engine that routes any refund under 50 USD for automatic approval. The chief operating officer asks which of the three is generative AI. Which response is correct?

   • The product description writer, because it produces new content in response to a prompt.
   • All three, because each one uses a model that learns patterns from company data.
   • The refund routing engine, because it makes decisions automatically without a person reviewing each case.
   • The demand forecasting system, because it was trained on historical data and produces numbers the company did not have before.

2. A regional bank has collected 30 candidate AI ideas from across the business during an internal campaign. The executive sponsor wants the first three initiatives to be chosen in a way the board will accept. The strategist has been asked to define the screening criteria before any idea is discussed on its merits. Which screening approach should be used?

   • Score each idea on business outcome, feasibility, and fit with the bank's stated objectives.
   • Rank ideas by how novel the underlying technology is, so the bank builds capability on the newest approaches first.
   • Choose the ideas that require the least data preparation so the first results arrive quickly.
   • Select the ideas submitted by the business units with the largest share of the bank's revenue.

3. A bank has deployed an AI model that recommends whether to approve small business loan applications. Applicants who are declined have a legal right to be told why. The model is accurate and well monitored, but the vendor describes its internal reasoning as proprietary and provides only a score with no factor breakdown. Which responsible AI dimension is most directly at risk?

   • Safety, because an incorrect decline could cause financial harm to a small business.
   • Privacy, because applicant financial data is being processed by a third party model.
   • Robustness, because the model may behave unpredictably on applications unlike those in its training data.
   • Explainability, because the bank must give declined applicants a reason for the decision.

4. A grocery chain's board has approved an ambitious AI agenda. The technology function reports that cloud infrastructure and data platforms are in place and that two data science teams are hired. The strategist has been asked to assess readiness before initiatives are launched, and finds that no executive owns the agenda, store managers have not been consulted, and there is no process for approving an AI system for use. What should the readiness assessment conclude?

   • The chain is ready, since infrastructure and technical talent are the binding constraints on AI delivery.
   • Readiness cannot be assessed until specific use cases are selected, since requirements differ by initiative.
   • The chain is not ready, since data quality has not been assessed and no readiness conclusion can be drawn without it.
   • The chain is technically ready but not organizationally ready on leadership, culture, and governance.

5. An insurance company wants to use AI to speed up claims triage. The data team reports that roughly 40 percent of claim records have a blank incident type field, and the free text adjuster notes use inconsistent abbreviations across three legacy systems that were merged after an acquisition. The program manager is preparing the business case for the steering committee. What is the most accurate business implication to present?

   • Missing fields are a technical concern for the data engineering team and do not change the business case.
   • Record volume matters more than record condition, so the project can proceed once the team confirms it has enough claims history.
   • Model outputs will carry the same gaps and inconsistencies, so data remediation belongs in the plan and budget before accuracy targets are set.
   • Free text adjuster notes are unstructured, so they should be excluded from the project scope.

6. A regional bank's credit team has been told that a new AI model outputs a probability rather than a yes or no decision. A senior manager asks why the system will not simply state whether an applicant should be approved, since that is what the team needs at the end of the process. Which explanation is correct?

   • The model returns a likelihood, and the bank sets the threshold that turns that likelihood into a decision.
   • The model is still in training, and it will produce firm decisions once enough applications have been processed.
   • Probability output indicates the model is uncertain, so the vendor should be asked for a more confident version.
   • The bank chose a generative model, and a predictive model would return the decision directly.

7. A B2B software company wants AI to improve sales performance. The sales director reports that win rates are healthy once a demo happens, but reps spend around 40 percent of their week researching accounts and writing outreach, and pipeline coverage is the constraint on revenue. Where should the AI investment be directed?

   • Sales forecasting, so leadership has a more accurate view of what will close each quarter.
   • Deal scoring, so reps prioritize the opportunities most likely to close.
   • Account research and outreach drafting, since that is where rep time is consumed.
   • Automated demo delivery, so more prospects can see the product without a rep present.

8. A gig economy platform uses a model to allocate delivery jobs to couriers. An internal review finds that couriers who decline jobs during school pickup hours receive fewer offers afterward, and that this pattern affects couriers with childcare responsibilities more than others. The model was never given any data about caring responsibilities. Which responsible AI dimension is most directly engaged?

   • Transparency, because couriers have not been told how job allocation decisions are made.
   • Privacy, because the model has inferred sensitive information about couriers' personal circumstances.
   • Fairness, because the allocation pattern produces systematically different outcomes for an identifiable group.
   • Robustness, because the model responds unpredictably to couriers who decline jobs.

9. A construction group scored well on a readiness assessment for data, infrastructure, and executive sponsorship. The one weak dimension was cultural preparedness: site managers described previous technology rollouts as things done to them, and two thirds said they expected the AI programme to be abandoned within a year like the last three initiatives. What does this finding mean for the programme?

   • It is a communication problem, and a launch campaign explaining the programme's benefits will address it.
   • Cultural readiness is a gating dimension, so the first initiative should visibly change that expectation.
   • It should be recorded as a risk and monitored, since culture changes slowly and cannot be addressed directly.
   • It reflects the previous initiatives rather than this one, so the programme should proceed on its own merits.

10. A logistics company is describing three systems to its board. The first flags shipments likely to arrive late based on ten years of delivery history. The second answers driver questions in natural language and drafts incident reports. The third checks whether a shipment weight exceeds a published limit and rejects it if so. The board wants each placed in the right category. Which categorization is correct?

    • All three are machine learning, since each one automates a judgment a person used to make.
    • The first is machine learning, the second is generative AI, and the third is rule-based automation.
    • The first two are both generative AI, since both produce an output the company did not previously have.
    • The first is generative AI, the second is machine learning, and the third is rule-based automation.

11. A vendor tells a hospital that its sepsis detection model is 97 percent accurate. The hospital's data shows that sepsis occurs in about 2 percent of admissions. The clinical director is impressed by the figure and wants to proceed. The strategist has been asked to review the claim before the contract advances. What should the strategist point out?

    • A model predicting no sepsis every time would score 98 percent, so accuracy alone says nothing here.
    • The accuracy figure was measured on the vendor's data and will be lower on the hospital's population.
    • Accuracy should be replaced by a measure of how confident the model is in each individual prediction.
    • The figure is meaningless without knowing how long the model took to reach each prediction.

12. A specialty materials company runs physical experiments to test formulations. Each experiment takes six weeks and costs about 40,000 USD, and the team runs roughly 30 a year from a candidate space of thousands of combinations. Selection is currently based on researcher judgment. Where does AI create the most value here?

    • Drafting the experimental write ups so researchers spend less time on documentation.
    • Automating the laboratory procedures so more experiments can be run in the same period.
    • Narrowing the candidate space so the 30 experiments target the best formulations.
    • Analyzing past experimental results to explain why previous formulations failed.

13. A rail operator is deploying an AI system that recommends track maintenance priorities. In testing, the system performed well on the conditions represented in its training data, but produced erratic recommendations when fed sensor readings from an unusually severe cold period the network had not previously recorded. Which responsible AI dimension does this engage most directly?

    • Fairness, since sections of the network in colder regions receive systematically different treatment.
    • Explainability, since the operator cannot determine why the recommendations became erratic.
    • Safety, since erratic maintenance recommendations could result in track failures and passenger harm.
    • Robustness, since the system behaves unpredictably on conditions outside those it was trained on.

14. A hotel group's executives all support AI investment. In interviews, the chief executive describes the goal as guest experience differentiation, the chief financial officer describes it as labor cost reduction, and the chief operating officer describes it as standardizing procedures across properties. Each believes the others share their view. What should the strategist do first?

    • Adopt the chief executive's framing, since guest experience is the most senior stated objective.
    • Design a portfolio with initiatives serving each of the three objectives, so every executive is supported.
    • Proceed with delivery, since the objectives are compatible and results will demonstrate value to all three.
    • Surface the three interpretations to the executives together and agree a single stated objective.

15. A vendor pitches a bank on what it calls a generative AI credit risk platform. In the technical session it emerges that the product scores applicants with a gradient boosted model trained on historical defaults, and uses a language model only to write a plain English paragraph explaining the score. How should the strategist characterize the product?

    • It is a generative AI platform, since the explanations customers receive are generated by a language model.
    • It is rule-based automation with an AI presentation layer, since gradient boosted models apply learned rules.
    • It is a predictive machine learning product with a generative explanation layer, and should be assessed as both.
    • The distinction does not matter commercially, since the bank is buying an outcome rather than a technique.

16. A retailer's customer service manager reports that the AI assistant keeps making the same mistake about a returns policy that changed in March. She says she has corrected it in the chat several times and expected it to have learned by now. What should the strategist explain?

    • The model does not learn from corrections in conversation, so the policy source or prompt has to change.
    • The manager needs to correct the assistant more consistently, since occasional corrections are outweighed by other conversations.
    • Corrections take effect after the model's next scheduled retraining, which may be several months away.
    • The assistant is retrieving an outdated policy document, which is the only possible cause.

17. A software company's engineering leadership wants AI to increase delivery throughput. Analysis of the last two quarters shows engineers spend 22 percent of their time writing code, 31 percent in code review waiting or reviewing, and 27 percent investigating production issues. Delivery is constrained by how long changes wait for review. Where should the first investment go?

    • Production issue investigation, since 27 percent of engineer time is reactive work that adds no new capability.
    • All three simultaneously, since the time is distributed and no single area dominates.
    • Code generation assistance, since writing code is the core engineering activity AI tools are built for.
    • Code review assistance, since review waiting is the largest block of time and the stated constraint on delivery.

18. A company plans an AI system that analyses employee communications to identify teams at risk of burnout. It would process messages from the internal chat platform and flag teams, not individuals. Human resources supports it, and the works council has not been consulted. Which concern should the strategist raise first?

    • Team level aggregation may not protect individuals in small teams, and staff were not consulted.
    • The system may misidentify at risk teams, producing interventions where none are needed.
    • The company should use survey data instead, since employees answer surveys knowing they are being measured.
    • Chat messages are unstructured data and would need substantial preparation before analysis.

19. A distribution group's readiness assessment scores four dimensions out of five. Leadership alignment scores 4, data quality 4, technical infrastructure 4, and governance 1. The programme director proposes reporting an average score of 3.25 and proceeding on the basis that the group is broadly ready. What should the strategist say about this reporting?

    • The average is a fair summary, since three of four dimensions are strong and governance can develop alongside delivery.
    • The assessment should be repeated, since a single dimension scoring 1 suggests the scoring was applied inconsistently.
    • Averaging hides a dimension that will block delivery, so the weakest score should be reported and addressed.
    • Governance should be weighted lower than the other dimensions, since it is a control rather than a delivery capability.

20. A distribution company's leadership team is reviewing two proposals. The first builds a dashboard showing last quarter's sales by region, product, and channel, with drill down. The second predicts which accounts are likely to reduce spending next quarter. The chief executive asks whether both are AI initiatives. What is the accurate answer?

    • Neither is AI, since both work from the company's own historical sales records.
    • The dashboard is business intelligence, and the prediction is machine learning.
    • The dashboard becomes AI once it includes forecast lines alongside the historical figures.
    • Both are AI, since both turn company data into insight that supports decisions.